Mitigating humanitarian crises during non-international armed conflicts—the role of human rights and ceasefire agreements

Full text



Open Access

Mitigating humanitarian crises during

non-international armed conflicts

the role

of human rights and ceasefire agreements

Lottie Lane


Background:Situations of humanitarian crisis are often caused by armed conflicts. Given the prevalence of non-international armed conflicts today, ways of ameliorating these situations are at the forefront of concerns. The international humanitarian law rules governing non-international armed conflict remain much less developed than those for international armed conflicts. This makes governing the behaviour of non-state parties to non-international armed conflicts (non-state armed groups) extremely challenging. This is exacerbated by the lack of direct human rights obligations for non-state armed groups. Although several initiatives have been taken to encourage non-state actors to mitigate situations of humanitarian crisis, the role of human rights law is somewhat under-developed in this respect. Method:The paper aims to assess what role human rights may have in improving humanitarian crises, suggesting one specific way in which they may be used. In order to achieve this, the paper will first discuss the international laws applicable to situations of non-international armed conflict. The paper will then critically analyse some of the initiatives that have already been taken to mitigate the disparity in human rights protection during armed conflicts.

Results:Based on the findings of this analysis, a new measure—the possibility of using a ceasefire agreement to impose specific human rights obligations on all parties to a conflict—will be assessed.

Conclusions: An evaluation of the merits and challenges raised by this suggestion will lead to a conclusion as to the role that human rights and ceasefire agreements could have during humanitarian crises.

Keywords:Humanitarian crisis, Economic, social and cultural rights, Non-state armed groups, Ceasefire agreements


The non-international armed conflict in Syria has been pro-claimed as causing the“biggest humanitarian emergency in our era”(Peralta 2014). The combination of a repressive re-gime, armed opposition groups and terrorist activities has left the country in a constant state of instability and chaos. Millions of people have been forced to live in abject condi-tions, without access to basic living supplies such as food, water and shelter. The ongoing grapple for power between the Islamic State and the Syrian government exacerbates the already dire situation. This paper will assess the poten-tial role of ceasefire agreements in imposing human rights obligations for non-state armed groups (NSAGs) in efforts to ameliorate situations of humanitarian crisis caused by non-international armed conflicts. The ultimate aim of the

paper is to build upon existing literature to suggest a way in which the inapplicability of human rights law to non-state armed groups could be circumvented during situations of humanitarian crisis in non-international armed conflicts (NIAC). The research question posed in this paper differs somewhat from the existing literature concerning the appli-cation of human rights law to NSAGs as it ties this discus-sion to the issue of humanitarian assistance.


The paper will first discuss the international law applic-able to situations of non-international armed conflict. This will involve a brief explanation of the doctrine of lex specialis, necessary for identifying the applicable law during non-international armed conflicts, and the stan-dards to which parties to the conflict may be held. The analysis will then focus on economic, social and cultural rights, as the most relevant for improving access to food, water and shelter. The section aims to demonstrate the

University of Groningen, Oude Kijk in‘t Jastraat 26, 9712 EK, Groningen, The Netherlands


aggravation of situations of humanitarian crises caused by the lack of direct human rights obligations for armed opposition groups. The next section of the paper will critically analyse some of the initiatives that have already been taken to improve inadequate human rights protec-tion during armed conflicts, made worse by the differ-ences in human rights obligations between non-state and state parties to a conflict. Based on the findings of this analysis, a new measure—the possibility of using a ceasefire agreement to impose specific human rights ob-ligations on all parties to a conflict—will then be assessed. This will include an evaluation of the merits and challenges raised by this suggestion, leading to a conclusion as to the role that human rights and ceasefire agreements could have during humanitarian crises.


Applicable rules of international law

The laws applicable to armed conflicts are extremely well rehearsed (Solis 2010; Clapham and Gaeta 2014) and will be only briefly laid out below. The following section will focus on non-international armed conflicts taking place between a state and (one or more) non-state armed group. For the purposes of the paper, the term ‘non-state armed group’ shall refer to a definition of a non-state actor as suggested by Geneva Call. It shall therefore include“any armed group, distinct from and not operating under the control of, the state or states in which it carries out military operations, and which has political, religious, and/or military objectives”(Bellal et al. 2011, 48).1

Applicability of international humanitarian law

The corpus of international humanitarian law applicable during NIAC is somewhat limited compared to that re-lating to international armed conflicts. In the early days of international law, the lack of development was per-haps due to a general understanding that because of its domestic nature, internal warfare fell within the scope of states’ national jurisdiction and need not be regulated internationally (Moir 2002, 2). Although somecustomary international law pertaining to NIAC existed (relating par-ticularly to the recognition of belligerency), state practice on the matter rapidly declined (Moir 1998, 352). However, as the prevalence of internal conflicts grew, and the trans-national effects of NIAC became more evident (e.g. the influx of refugees and/or a ‘spill-over’ of hostilities to neighbouring states: Moir 2002, 2) and the realisation that parties most affected by conflicts (i.e. civilians) were in need of protection regardless of the nature of the conflict, the mid-twentieth century brought a greater acceptance of the application of humanitarian norms to NIAC. None-theless, despite efforts of the International Committee of the Red Cross to encourage the application in practice (having adopted a resolution on the matter in 1938: Moir

1998, 354), progress was stopped short by the breakout of World War II. It was therefore not until 1949, after a re-jection of the International Committee of the Red Cross (ICRC)’s attempts to have the totality of international hu-manitarian law extended to cover NIAC, that the some-what restrictive Common Article 3 to the universally binding Geneva Conventions (Best 1994, 82–83; ICRC 1949a; ICRC 1949b; ICRC 1948c; ICRC 1949d) was adopted.2The NIAC-specificAdditional Protocol II to the Geneva Conventions of 12 August 1949(ICRC 1977) was later hastily adopted in 1977, after further disagreements be-tween states as to the extent to which their internal affairs should be opened to external scrutiny (see Best, 346–347).

The standards contained in these instruments apply to both state and non-state parties to a NIAC. This asser-tion has raised several conceptual concerns, perhaps the most notable relating to the legitimacy of applying treaty standards to NSAGs, who have not ratified the relevant treaties. At the international level, in the absence of an elected world government, the legitimacy of obligations stems from the sovereign equality of states and the fact that they bind only themselves through the creation and adoption of international norms. The source of legitim-acy for the imposition of direct obligations on non-state actors at the international level without their participa-tion therefore raises some quesparticipa-tions. Justificaparticipa-tions prof-fered range from the ‘doctrine of state jurisdiction’3 to the analogy of individual criminal responsibility.4 Per-haps the most persuasive justification is the argument that some procedural requirements of legitimacy need not be followed in relation to norms preventing heinous conduct. It has been argued, for example, that if the ex-pected result of the obligations’ implementation is of paramount importance, it may negate the necessity of the norms being adopted with the consent of affected parties (Ryngaert 2010, 71–73). Ryngaert asserts that in the absence of participation by a non-state actor, if a “legal norm or its implementation has in itself an im-portant substantive value”, participation is not necessary (Ryngaert 2010, 71). Arguably, in the case of inter-national humanitarian law that was extended to NIAC primarily for the purpose of protecting civilians, this ar-gument rings true. Indeed, “it has now become uncon-troversial…that [NSAGs] are bound by international humanitarian law”(Bellal et al. 2011, 56).


been more broadly accepted than the application of treaty-based rules (Bellal et al. 2011, 56).

In addition, the assertion that some rules of inter-national armed conflicts are also applicable in NIACs is becoming more commonplace (Columbia Law School Human Rights Institute, cited in Cleveland 2014). These evolutions in the law of NIACs have taken place pre-dominantly to counter concerns that despite the wide-spread nature of NIACs, the laws applicable during an international armed conflict were much more extensive in scope. Until the 1990s, developing the rules of NIAC beyond those provided for by Geneva law was “never seriously entertained” (Sivakumaran 2012, 55). With the majority of armed conflicts currently occurring world-wide being non-international in nature (Armed Conflict 2014),6the developments are now a welcome opportun-ity to mitigate the human suffering caused by armed conflicts and thwart concerns regarding the deregulation of NIACs.7

Now that the law applicable during NIAC is in a more (though by no means fully) developed state, the primary issue to be addressed is how to ensure that NSAGs com-ply with the relevant norms and close the gap between law and practice during NIAC. Aligning the practice of NSAGs with the legal standards is an extremely challen-ging task, however. On the one hand, NSAGs may not be aware of the existence or meaning of humanitarian norms and may lack the institutional structure to ensure compliance of their own fighters (Ryngaert and Van de Meulebroucke 2012, 456–457). Alternatively, a NSAG may be unwilling to engage with the notion that it has legal obligations under humanitarian law (e.g. ICRC 2008). Even those NSAGs willing to abide by the laws may encounter an array of obstacles in implementing them.8It is for this reason that organisations such as the International Committee of the Red Cross have engaged with NSAGs that have expressed a desire for help. Nonetheless, as Ryngaert and Van de Meulebroucke note, unlike for states, there is no formal advisory ser-vice available to NSAGs struggling to comply (Ryngaert and Van de Meulebroucke 2012, 457).

Notwithstanding the difficulties faced in ensuring NSAGs’ compliance with humanitarian law, there is growing support for the application of international hu-man rights law to situations of armed conflict, which will be explained in the following section.

Applicability of international human rights law

The rhetoric pertaining to the application of inter-national human rights law to armed conflicts was ini-tially somewhat divergent. A major focus of this debate has revolved around the doctrine of lex specialis derogat legi generali(lex specialis), which many believed to ren-der the application of human rights during armed

conflicts inappropriate (e.g. Schabas 2007, 594). As the lit-erature engaging with this debate is extremely extensive, this section will present a summary of the doctrine and its consequences within the context of this paper. The doc-trine mandates that more precise and specialised law is to take precedence over more general laws. Reluctance to apply human rights law during armed conflicts was also due to the differing natures and ‘roots’ of humanitarian and human rights law (Solis 2010, 24). On the one hand, humanitarian law (not seeking to abolish conflict) does not prohibit the immediate killing of the enemy. It seeks rather to diminish the devastating human cost of conflicts and to ensure a fairer fight (ICRC 2004).9Human rights law, on the other hand, seeks to protect individuals from potentially abusive actions by states and does not allow the arbitrary taking of life under any circumstances.10

Furthermore, as is explained below, human rights law imposes obligations on the state for the benefit of indi-viduals. The relationship between the obligation-holder and the beneficiaries (or rights-holder) is thus inherently vertical in nature. In contrast, as explained above, many humanitarian obligations are owed by all parties to the conflict, which act essentially as mutual beneficiaries (assuming that all parties comply with their obligations).

In more recent years, the international community has increasingly accepted the view of the International Court of Justice (ICJ) in its Advisory Opinion on the Legality of the Construction of a Wall in the Occupied Territory that during armed conflicts international humanitarian norms and international human rights norms may apply simultaneously, in a complementary manner (Legal Conse-quences of the Construction of a Wall in the Occupied Pal-estinian Territory, Advisory Opinion 2004, para 106). It may be said, therefore, that the doctrine of lex specialis serves more to determine the preciserules to apply to a particular situation, rather than precluding application of one body of law. This view is supported by Marko Milano-vić, who has highlighted that understanding the doctrine as being generally applicable to the human rights and hu-manitarian regimes as a whole, is mistaken (Milanović 2011, 98–101). Following Kreiger, Milanović’s suggestion is to assess which rule constitutes thelex specialisby looking at the relationship between specific norms, rather than re-gimes as a whole (Milanović 2011, citing Kreiger 2006). The present paper will also depart from this starting point, understanding the lex specialis during situations of hu-manitarian crisis as being human rights law.

Economic, social and cultural rights


thelex specialisduring armed conflicts are civil and pol-itical rights, such as the right to life and the prohibition of torture. This is perhaps due to the existence of con-crete norms in humanitarian law which also provide rules on the use of torture and the taking of life,11 al-though they differ from human rights law. However, the rights affected (and applicable) during armed conflicts are not limited to those whose subject matter is also dealt with by norms under humanitarian law. When concerned with humanitarian crises that result in heavily reduced access to materials and services essen-tial to a life of dignity, often referred to as ‘subsistence rights’(such as healthcare, food, and water and sanita-tion), economic, social and cultural rights are of the utmost relevance.

It may well be argued that in relation to the provision of food and water during armed conflicts, human rights law constitutes thelex specialis. As implied above, deci-phering which norm/s form the lex specialis in a given circumstance will require an examination of which norms are the most developed. In the present context of subsist-ence rights, human rights law has not only been given more content than the relevant humanitarian law norms but also provides (in theory) more extensive protection of access to essential materials and services.

For example, humanitarian law rules do prohibit the use of starvation as a method of warfare and the targeting of essential resources (being classed as civilian objects),12 thereby providing limited protection of materials. The rules on access to humanitarian aid are more developed for international armed conflicts. For non-international armed conflicts, however, the applicable treaty rules do not explicitly refer to humanitarian aid (Swiss Federal Department of Foreign Affairs 2011, 25–26). Regardless, the Swiss Federal Department of Foreign Affairs has inter-preted Common Article 3 to include a principle that civil-ian populations may not be intentionally subjected to situations that would, due to a lack of access to essential supplies, threaten their dignity or result in“serious mental or physical suffering” (Swiss Federal Department of Foreign Affairs 2011, 26). However, these rules are con-structed as ‘negative’ obligations—prohibitions of certain conduct requiring parties to refrain from interfering with access to essential supplies. Whilst the same obligations can be found under economic, social and cultural rights, human rights law goes further, requiring states parties to not only respect the rights (i.e. refrain from interfering with their enjoyment) but alsoprotectandfulfilthe rights by providing the means and/or substance for the right to be effectively realised (Shue 1980, 260; Eide 1987).13

In addition, the rule in Common Article 3 relating to the lack of access to essential supplies is not buttressed by a wider range of provisions applicable during NIAC. Indeed, humanitarian assistance (i.e. the provision of

food, water and healthcare) as such is scarcely regulated during NIAC, which may raise a presumption that hu-man rights law constitutes the lex specialis. Provisions regulating humanitarian assistance during international armed conflicts can be found in (for example) Article 23 Geneva Convention IV and Article 70 Additional Protocol I (see ICRC, 2014). However, other provisions relating to humanitarian assistance during NIAC are limited to Article 18 Additional Protocol II. This article provides for the undertaking of humanitarian“relief actions”(with the consent of the concerned state party) in the event that the “civilian population is suffering undue hardship owing to a lack of the supplies essential for its survival” (Article 18, para 2). The very vagueness of this protection, extending to “foodstuffs and medical supplies” suggests that the much more embellished human rights law would offer more protection of subsistence rights than humanitarian law. It may be argued that theManual on the Law of Non -International Armed Conflict 2006 (the San Remo Manual) provides further detail in this respect. The manual is a restatement of the law applicable during NIAC, though like the ICRC codifications of law, it is not legally binding. Chapter 5 of the Manual states that “humanitarian assistance should be allowed and facilitated by those engaged in military operations whenever essential needs in an emergency are not being met” and provides more information on the definition of humanitarian assistance. The Manual still refrains from placing positive obligations on par-ties to the conflict to ensure that the essential needs are, in fact, met. Taken together with the lack of more de-tailed information on what constitutes essential foodstuffs etc., this contributed to the argument that thelex specialis in the present context is human rights law.

Further support for the argument that human rights law constitutes thelex specialisin relation to a humani-tarian crisis caused by a NIAC may be found in the non-limitation and non-derogability of subsistence rights in such a situation. The implications of these two aspects of subsistence rights and the nature of the obligations deriving from them, as economic, social and cultural rights, will now be explained.

Limited application of economic, social and cultural rights

A. Legitimate limitations of human rights


Legitimate limitations to economic, social and cultural rights are allowed under Article 4 of the International Covenant on Economic Social and Cultural Rights (ICESCR) for the promotion of the general welfare in a democratic society, providing that they are not contrary to the nature of the right. This sole reason justifying limitations is more restrictive than the several reasons found in the

International Covenant on Civil and Political Rights 1966a(ICCPR). Article 19 ICCPR on freedom of expression, for example, allows limitations for several reasons, including the respect of the rights or reputation of others, the promotion of national security or pubic order, or of public health or morals. Müller has persuasively argued that this reason effectively means that states may not limit the‘minimum core’of economic, social and cultural rights, since they would go against the nature of the rights (Müller2009, 575).15In addition to these requirements, limitations to economic, social and cultural rights must be prescribed by law,

proportionate to the aim pursued, and necessary in a democratic society. Furthermore, as the Limburg Principles on the Implementation of the International Covenant on Economic, Social and Cultural Rights state,“[article 4] was not meant to introduce limitations on rights affecting thesubsistence…of the person.”(United Nations Commission on Human Rights1987, para 47). Taken together with the nature of subsistence rights as providing for the basic necessities required for human existence, it can therefore be inferred that limitations to subsistence rights would not be considered legitimate.

B. Derogations from economic, social and cultural rights The question of whether state parties may derogate from economic, social and cultural rights in times of public emergency has been a matter of much debate. Derogating from a right essentially allows states to put their obligations on hold for a specified period of time. As derogating is a somewhat extreme measure, whether or not a particular right may be derogated from, and under which circumstances, is usually laid down in the text of a human rights treaty. However, this is not the case for the ICESCR, which neither contains a derogation clause allowing for derogations, nor a provision prohibiting

derogations. This is unlike the ICCPR, Article 4 of which specifies the conditions for derogations from its provisions, and prohibits derogations from several rights (Article 4, paragraphs 1 and 2, respectively). Nonetheless, the fact that there is no derogation clause in ICESCR does not necessarily mean that states would be precluded from derogating from them.

However, itcanbe inferred from the purpose of derogation clauses that at least some economic, social and cultural rights are non-derogable. According to Müller, this would extend to

subsistence rights (Müller2009, 593). The purpose of derogations is not to allow states to decrease their attention to the rights, but (following the criteria of Article 4 ICCPR) must be to ensure that the state is in a position where it is capable of ensuring human rights and to restore a situation of normalcy (United Nations Human Rights Committee2001, para 1). This is evident from the requirement that a state be in a“time of public emergency that threatens the life of the nation”before it may make derogations. Whilst it may be true that a non-international armed conflict may cause such a situation of public emergency, it cannot reasonably be expected that derogating from rights such as the right to food, water and healthcare, could help to restore the state to a situation of normalcy. On the contrary, reducing access to essential resources would aggravate, rather than ameliorate, a situation of public emergency. Allan Rosas and Monika Sandvik-Nylund have also suggested that the relationship between subsistence rights and the right to life can contribute to the argument in favour of the non-derogability of subsistence rights (Rosas and Sandvik-Nylund,2001, 414). Subsistence rights are of the utmost importance for the protection of human dignity and survival in emergency situations and are interrelated with the right to life—a non-derogable right (to the extent that a life may not be arbitrarily taken) (see Müller2009, 599). This view is supported by several human rights bodies which, lacking jurisdiction over (or the justiciability of) economic, social and cultural rights, have interpreted the right to life to include subsistence rights. For example, the Inter-American Court of Human Rights has repeatedly read the right to life (protected by Article 4 of the American Convention on Human Rights) to include healthcare as one of its essential attributes (see Case of Children´s Rehabilitation v. Paraguay 1999). This reading is now “solidly part” of the Court’s jurisprudence, having been embellished upon in several cases (e.g. Yakye Axa Indigenous Community v. Paraguay 2006, discussed in Melish 2008, 389). C. The nature of ESCR

Whether or not the manner of using the right to life in this way is found to be persuasive for present purposes, the fact that Article 2(1) ICESCR allows economic, social and cultural rights to be

progressively realised is also of relevance here.16


the realisation of economic, social and cultural rights, their full realisation is not an immediate obligation. States must, however, make continuous and

progressive measures to increase the realisation of the rights, depending on their available resources. Whilst it may be argued that a state has less resources available during armed conflicts, which would naturally lead to a lesser degree of the rights’realisation, the Committee on Economic, Social and Cultural Rights (CESCR) has introduced a concept of‘minimum core obligations’of the Covenant rights (CESCR1990, para 10). Simply speaking, this means that there is a certain floor of human rights realisation that states must ensure, regardless of their particular domestic situation. In relation to the rights to food, water, and highest attainable standard of health, which have been given more content through CESCR’s General Comments 12, 15 and 14, respectively, the minimum core would arguably provide more protection of subsistence rights than the norms under humanitarian law, despite their progressive nature.

In addition, the Committee has suggested that the notion ofprogressiverealisation makes it extremely cumbersome for states to justify any retrogressive measures.17

The extent to which this would also hold true during armed conflicts, during which time resources may need to be redistributed, is unclear. However, it may be deduced that at least the minimum core of subsistence rights may not be derogated from, even during situations of public emergency threatening the life of the nation. This conclusion is particularly significant when viewed in light of the below discussion on the absence of human rights obligations for NSAGs. If subsistence rights were derogable, it would mean that the obligations on the state and NSAGs would be more equal and would provide some level of justification for the fact that individuals’rights were not being realised.

The above discussion demonstrates that the more embellished standards relating to the provision of food, water and healthcare found within international hu-man rights law makes these norms, rather than those found in humanitarian law, the lex specialis in the present context. This finding is strengthened by the conclusion that subsistence rights are non-derogable and may not be limited during NIAC. Unfortunately, whilst this affords perhaps more protection to individ-uals within territory controlled by a state, it leaves in-dividuals in areas controlled by NSAGs (as non-human rights obligations-holders) without human rights pro-tection. This problem and efforts to use human rights law to remedy the gap in protection through the indir-ect application of human rights obligations to NSAGs will now be assessed.

NSAGs and the indirect horizontal effect of human rights Under the current international human rights framework, NSAGs do not have direct human rights obligations; as non-state actors, NSAGs cannot be party to international human rights treaties. Rather, human rights obligations are vertical in nature, being owed by the state on a vertically su-perior legal plane, to individuals. Unlike other international treaties, human rights treaties do not even have horizontal application, or ‘horizontal effect’ between states. Human rights treaties allow for neither substantive horizontal effect (i.e. allowing individuals to claim violations of actual rights owed to them by NSA), nor procedural horizontal effect, often known as drittwirkung18 (reflected in the rule that complaints of human rights violations may only be brought before the human rights treaty bodies by individuals against states) (e.g.Optional Protocol to the International Covenant on Civil and Political Rights1966c, Article 1).

However, the ever-increasing power and influence of non-state actors (including NSAGs) and their ability to interfere with the enjoyment of human rights has led to an application of ‘indirect horizontal effect’ by many hu-man rights monitoring bodies and courts. Although being achieved through several different methods, indirect hori-zontal effect essentially holds states responsible for human rights interference resulting from the conduct of non-state actors. In the past, this has been used in relation to corporations, NSAGs, and even individuals (e.g. European Court of Human Rights (ECtHR)Fadeyeva v. Russia2005; United Nations Committee against Torture Sadiq Shek Elmi v. Australia 1999; ECtHR, X and Y v. The Netherlands 1985). Perhaps the most prevalent way of achieving indirect horizontal effect is to use states’positive obligation to protect individuals from human rights inter-ference with the enjoyment of their rights.

The obligation has been fleshed out using a‘duty of due diligence’. Under international human rights law, this means that states must take all appropriate measures to ‘prevent, investigate and punish’ third party interference with human rights enjoyment (Inter-American Court of Human Rights Velásquez-Rodríguez v. Honduras 1988, paras 79, 172). When assessing whether or not a state has acted with due diligence, the focus is on the conduct, and the progressive measures that states took in a particular situation, rather than theresultof the measures and whether or not they were actually successful in protecting an individ-ual (Office for the United Nations High Commissioner for Human Rights 2005, 61). In practice, this means that a state will not be expected to protect every individual all of the time. Rather, they will be expected to take protective mea-sures when they know, or should have known, of a threat to an individual’s rights by a third party (see, e.g. ECtHR, Osman v. United Kingdom1998b; ECtHR,Z and Others v. United Kingdom2001, ECtHRKaya v Turkey1998a).


of state control, such as NSAGs. An illustrative example is that of the Fuerzas Armadas Revolucionarias de Colombia (FARC) in Colombia. The non-international armed conflict between the NSAG and the state has been taking place for over 50 years, with the FARC hav-ing gained in 2000 effective control over a large area of Colombian territory (Rollins 2010, 18). The Human Rights Council has expressed concern at the lack of inquiry and investigation into crimes committed by demobilised individuals from the FARC against women and children, in particular the recruitment of child sol-diers (Human Rights Council, Working Group on the Universal Periodic Review 2008, para 57).

This would also presumably fail to comply with the state’s obligation under Article 4(2) of the Optional Protocol to the Convention on the Rights of the Child on the Involvement of Children in Armed Conflict(2000) to “take all feasible measures” to ensure that the relevant norms are respected. The Inter-American Commission on Human Rights emphasised the importance of the ob-ligation to investigate and punish actions by NSA in its assessment of whether Colombia had acted with due diligence in relation to FARC activity, but ultimately stated that “in situations of civil strife, the State cannot always prevent, much less be held responsible for, the harm to individuals and destruction of private property occasioned by the hostile acts of its armed opponents.” (Inter-American Commission on Human Rights, Organization of American States 1999, Chapter IV para 4, discussed in Clapham 2006b, 422). This appears to place a lower (albeit more realistic) burden on states than the European Court of Human Rights. In the case ofIllascu and others v. Moldova and Russia(1999), the Court was called upon to question the responsibility of Moldova for harm that occurred in an area of its territory over which it no longer had effective control. The Court opined that “States retain the obligation to use all means and re-sources available to them to guarantee human rights” (para 333) and upheld Moldova’s responsibility. Whilst en-couraging states to make efforts to guarantee human rights throughout its territory regardless of situations of conflict is laudable, finding the state to have violated its obligations in areas where it is no longer capable of secur-ing human rights is questionable. Although the approach of the Inter-American Commission may appear to be too soft-handed, the much more heavy-handed approach of the ECtHR has been questioned, not least by dissenting judges (Illascu and Others v. Moldova and Russia 1999, Partly Dissenting Opinion Of Judge Sir Nicolas Bratza).

Ultimately, whichever view is taken could result in a gap in human rights protection. Even if a state were to use all means and resources available to try to secure human rights in areas controlled by NSAGs, it may not be possible. Additionally, and unfortunately, the vast ma-jority of previous cases upholding indirect horizontal

effect have been in relation to civil and political rights. Until recently, it was not possible to bring an individual com-plaint in relation to rights contained in the International Covenant on Economic, Social and Cultural Rights.19The entry into force of the Optional Protocol to the Covenant now allows for this possibility, but it remains to be seen how the Committee on Economic, Social and Cultural Rights will deal with such situations.20These factors all cul-minate in a gap in effective legal protection of subsistence rights during armed conflicts. Whilst some NSAGs take it upon themselves to provide public services and to essen-tially fulfil some human rights on ade factobasis, there ex-ists a legal lacuna. A correlative of this is an inequality in human rights protection. Victims living in an area con-trolled by the state may still be able to receive redress for their human rights violations by bringing a complaint dir-ectly against the state. For those living in NSAG-controlled areas, depending on the situation on the ground and the ef-forts that states have made in securing human rights enjoy-ment despite the control of the NSAG, this may not be possible. Individuals suffering the effects of severe humani-tarian crises may therefore be left with no way of accessing essential materials. Despite laudable efforts by humanitarian aid organisations to deliver materials to those in need, and the humanitarian norms prohibiting the restriction of their access to areas in need of essential materials,21some areas remain rife with crisis. For these reasons, more measures need to be taken to try to achieve a rounder, more compre-hensive protection of human rights.

This contrasts with international humanitarian law, which contains the fundamental principle of equality of obligations.22 This means that all parties to a conflict owe the same obligations and hold the same rights“ irre-spective of the ‘justness’ of the cause”, even during NIAC (Sivakumaran 2012, 242–246). Consequently, ci-vilians belonging to both sides of the conflict are in the-ory equally protected from the effects of the conflict. However, the developments in the range of laws applic-able during NIAC remains limited in comparison with international armed conflicts due to the reasons men-tioned above. The limited scope of the norms renders the equality of obligations during NIAC less meaningful, particularly in situations where the high threshold for application of Additional Protocol II is not met. Indeed, the equality of obligations in NIAC was seemingly a re-sponse to the need to ensure equal protection for civil-ians during internal as well as international conflicts rather than to recognise NSAGs as bodies competent of discharging obligations.


gaps in the obligations of state vs. non-state actors. With respect to both legal fields, the primary reason for this is the prevailing state-centric, Westphalian approach to international relations. States are still considered to be the sole subjects of international law; sovereign entities endowed with the power and responsibility of managing their internal affairs (including the regulation of non-state actors). This is reflected, for example, in the fact that only states may be party to international human rights law treaties and the Geneva Conventions (including Additional Protocol II). Changes in the prevalence and power of NSAGs are highlighting the insufficiency of this paradigm for dealing with situations of humanitarian crisis during NIAC. Interestingly, some level of equality has been trans-posed into the younger field of international criminal law, which allows for individuals to be held individually crimin-ally responsible for breaches of international humanitarian law. This is possible where an individual has committed one of the ‘grave breaches’ of the norms, some of which concern human rights principles.23Nonetheless, the princi-ples involved do not relate to subsistence rights. This makes the deterrent aspect of international criminal law less valu-able in the context of humanitarian crises. Consequently, the gap in both humanitarian and human rights law makes it difficult to govern the actions of NSAGs effectively.

Christa Rottensteiner has noted, “the primary respon-sibility for meeting the needs of the civilian population in an armed conflict rests with the warring parties that are in effective control of the territory on which that population lives” (Rottensteiner 1999). As the inter-national legal framework does not currently envisage NSAGs as obligation-holders, and in light of the lack of evidence of political will to alter this, extra-legal efforts are necessary to bring NSAGs’conduct in line with hu-man rights standards and alleviate huhu-manitarian crises caused by NIAC.

Methods for improving the governance of NSAGs There have been numerous methods used to reduce the human cost of non-international armed conflicts, many of which also aim to more effectively govern the actions of NSAGs. The measures range from reports condemning the actions of the groups to voluntary undertakings by groups promising to adhere to particular international norms. The scope of the measures’contents is also vari-able, seemingly rendering specific examples more or less relevant to the issue of improving the fulfilment of subsist-ence rights during non-international armed conflicts. However, the common aims of each measure (being to en-gage with NSAGs and bring their conduct into compli-ance with particular legal standards) make each example discussed below relevant in the present context. Several previous initiatives will now be assessed to determine the likelihood of similar approaches being able to improve the

protection of subsistence rights during situations of hu-manitarian crisis.

Voluntary undertakings

As can be seen on quite a widespread basis, NSAGs with effective control over an area of territory sometimes volun-tarily undertake some human rights obligations. Whether the group itself views their actions in this way or not, there are many which have, on a de factobasis, begun to fulfil some human rights. These can range from the provision of some public services to the instatement of an internal just-ice system. This is often due to the group’s ultimate goal of either taking complete control over a territory and becom-ing the new governmental authority or establishbecom-ing a sep-arate, smaller state within the territory of the state that they are in a conflict with. We can see examples of both of these instances if we look at the so-called Islamic State and the FARC, respectively. The mission of the Islamic State is to take control over a very large territory within the Le-vant, including Iraq and Syria (Australian National Secur-ity Database on Terrorist Organisations 2014). Within a relatively short period of time, the group gained effective control over an area of Syrian territory, establishing a ‘capital’ known as Raqqa (Hubbard 2014). Having ob-tained authority in this region, the Islamic State went on to establish what is essentially a state-like structure (Barnes-Dacey 2014). This has involved the group re-securing the provision of some public services, for ex-ample installing new power lines and setting up a‘suq’ for locals to exchange goods, and reforming the educa-tion system (see Zelin 2014). Whilst the inteneduca-tions of the group in doing these are most likely not related to human rights concerns, it could be argued that they are in effect providing some human rights to (at least some) individuals within Raqqa. However, obvious problems arise. There are concerns as to discrimination in the provision of the public services, and the inability of the Syrian state (or indeed of third states) to exercise con-trol or influence over the group makes it virtually im-possible for outside actors to improve the situation of humanitarian crisis within Raqqa.

The situation regarding the FARC is somewhat differ-ent. In 2000, the Colombian Government agreed to con-cede around 42,000 km2 to the non-state group. Some of the actions of the FARC within this area have been surprisingly state-like, as the group“deliver[s] social ser-vices, including healthcare and education. They also practise restorative justice through their revolutionary courts and have implemented agrarian reform by break-ing up large ranches and turnbreak-ing over smaller plots to landless peasants. They also collect taxes from local businesses to fund schools and clinics”(Leech 2013).


are providing ‘public’ services such as education and healthcare suggests that they are capable of and actually are fulfilling these human rights within their controlled territory.24 The successful governance of at least this relatively large area by the FARC makes it not implaus-ible that they would be able to orchestrate the required infrastructure to achieve this on a larger scale, were their armed forces be able to overtake a larger percentage of land from the state.

Voluntary undertakings in this manner are a promising development for subsistence rights. Many public services entail the provision of economic, social and cultural rights, such as healthcare. However, the nature of the undertak-ings makes them very hard to regulate and monitor. The lack of a concrete agreement or obligation means that the NSAG providing the services may choose the extent to which it wishes to provide a right. As with ISIS, this may in turn lead to discrimination in the fulfilment of rights. It is therefore desirable to have some form of agreement or statement by the NSAG specifying the scope of the rights and obligations they are willing to fulfil.

Action plans and deeds of commitment

There have been several initiatives taken by the United Nations and various non-governmental organisations to encourage NSAGs to adopt agreements specifying obli-gations with which they agree to comply. Most of the measures focus on humanitarian norms, rather than hu-man rights law. This is logical, since NSAGs are subject to some humanitarian obligations, but the lack of pres-sure on groups to respect human rights norms dis-tinctly from humanitarian norms could be a missed opportunity. Two of the largest initiatives taken to bet-ter govern the actions of NSAGs are the Security Council action plans and lists of shame (within the context of the recruitment and use of child soldiers) and Geneva Call’s Deeds of Commitment. Lessons may be learned from these two examples as to the likely challenges to securing a human rights-specific under-taking by NSAGs. Indeed, experiences with the action plans and deeds of commitment can demonstrate whether a solution to the problem at hand (securing NSAGs’compliance with subsistence rights to alleviate humanitarian crises) may be found in these two measures.

A. Action plans

Action plans were introduced after the General Assembly’s Special Representative had identified six

“grave violations”towards children during armed conflict. The plans are formed following Security Council action in this context (see Office of the Special Representative of the Secretary-General for Children and Armed Conflict2009, 2).

In 2001, the Security Council adopted a resolution which urged the Secretary-General of the United Nations to publish a list of all parties to armed conflicts who were recruiting or using child soldiers incompatibly with applicable obligations under international law, and in relation to situations which were, or could have been, on the agenda of the Security Council (see United Nations Security Council2001, para 16). As a consequence, with the intention of

“naming and shaming”armed forces, the first‘1379 list’ in 2002 contained 23 groups (Secretary General on Children and Armed Conflict2002, cited in Harvey

2003, 29–30). A further resolution now obliges to provide reports to the Security Council on actions taken to ameliorate the situation (United Nations Security Council2003, para 5, cited in Harvey

2003, 30).

A further initiative by the Security Council—the


the need to take further action to relieve situations of humanitarian crisis.

B. Deeds of commitment

Similar comments can be made in relation to Geneva Call’s Deeds of Commitment. There are three types of commitment, dealing with anti-personnel mines, the protection of children from the effects of armed conflict and the prohibition of sexual violence and gender discrimination (see Geneva Call2014a,2014b). The deed relating to the protection of children has been a landmark

development within the global campaign against child soldiers. The deed is the first international instrument that NSAGs could voluntarily and unilaterally sign and be judged upon their implementation thereof. Nevertheless, they may only have real value as a positive gesture. The deeds have been instrumental in raising awareness and encouraging NSAGs to consider the human rights impacts of their actions more concretely. In particular, the deed for the prohibition of sexual violence and gender discrimination provides a substantial list of commitments to which the groups agree to adhere to. These include certain provisions that would also fall under international human rights law, such as a prohibition of discrimination against women and equal access to healthcare (Geneva Call2013a, para 5). Additionally, the deed does acknowledge that it is“one step or part of a broader commitment”to human rights and humanitarian law (Geneva Call2013a, para 9). Nonetheless, there is no direct reference to human rights obligations of NSAGs.

In contrast, the deed on the protection of children does mention human rights in its main provisions, but restricts commitment to respect for the rights to life, human dignity and development (Geneva Call

2013b, preamble).26It is of course understandable that Geneva Call refrained from including a fuller range of human rights in the deed. Given the worries of legitimising NSAGs by holding them to the same international obligations as states during armed conflicts (which has also contributed to the fact that rules of non-international armed conflicts are less expansive than those of international armed conflicts) (Sivakumaran2012, 68–77), concerns that NSAGs do not have the capacity to fulfil human rights obligations to the same extent as states (e.g. Moir2002, 194, cited in Clapham2006a, 502) and the aim of the deeds, the focus on humanitarian norms is not misplaced. Nevertheless, as per the approach taken by the Inter-American Court (discussed above), which would read some economic, social and cultural rights into the right to life, the

deeds could be interpreted to impose obligations on NSAGs to contribute to the realisation of subsistence rights. But even if the deeds were to be read as such, two main problems ensue. On the one hand, the commitment is to‘respect’the rights, which under international human rights law is an obligation to refrain from interfering with the enjoyment of human rights. Individuals in situations of humanitarian crisis, however, require their rights to befulfilled. This obliges obligation-holders to (i) facilitate the realisation of rights by taking“positive initiatives to enable the full enjoyment”and (ii) provide“direct or indirect state services when individuals or groups are unable, for reasons beyond their control, to realise the right themselves by the means at their disposal”(see respectively, CESCR1999a, para 15; CESCR1999b, para 6). This goes considerably beyond an obligation to refrain from taking action and cannot legitimately be read into a commitment to‘respect’rights. The other problem is the non-binding nature of deeds and the lack of any enforcement mechanism for them, which may not result in concrete improvement of victim’s right, and the link between the deed and obligations of subsistence rights is a somewhat tenuous basis for imposing a direct obligation on NSAGs.

C. Outcomes of the initiatives


to ensure that they are implementing the agreements (Geneva Call2015a). This is an important step in being able to fulfil the deeds’goals of holding NSAGs publicly accountable for their actions (Geneva Call2015b). For a NSAG, signing a deed of commitment is one step towards acknowledging (albeit limited) international responsibility for its actions. The deeds suggest that once engaged, NSAGs are willing and capable of taking commitments seriously. If groups are hoping to establish themselves as a legitimate authority, it is crucial for them to be seen to make a tangible effort to abide by international obligations to which they would be bound were they to succeed. This is all the more important in light of theArticles on the Responsibility of States for Internationally Wrongful Acts(International Law Commission2001), according to which an insurrection group that succeeds in becoming the legitimate authority of a state may be held,ex post facto, internationally responsible for any breaches of international law that may be attributed to it before it came into power (Article 10).

However, notwithstanding great expectations being placed on United Nations action to combat the use of child soldiers, reports show that the technique of naming and shaming has not been successful. Instead of decreasing the prevalence of child soldiers, the most recent report by the Secretary-General on children and armed conflict states that there are currently 57 groups in 15 countries that are recruiting or using child soldiers,29

17 of which have remained on the list for 5 years or more (Coomaraswarmy2012, Annex 1). In contrast, only five groups were‘de-listed’in the last year.30In addition, the prohibition of the recruitment and use of child soldiers by NSAGs in Article 4 to the Optional Protocol to the CRC does not afford as effective protection as may be hoped. The lack of monitoring or enforcement mechanisms within the Protocol“inevitably will hinder”its effectiveness (Abraham2003, 17). This is particularly true in relation to NSAGs, which unlike states are not subject to external monitoring mechanisms (e.g. the Human Rights Council’s Universal Periodic Review). This strongly supports an argument that there need to be supplementary measures taken in the area of human rights; the current legal regimes, whether within human rights law or in other areas of international law and policy, are clearly not succeeding in their goals.

In the context of the current problem, the action plans and deeds may appear to be of limited value. However, in terms of moving forwards, Andrew Clapham has suggested extending the commitments

to cover more detailed human rights abuses (Clapham2006b, 292–293). On a similar vein, Solimon Santos has envisaged Geneva Call basing these commitments on human rights in the future (Santos2002, cited in Clapham2006b, 293). This could indeed be useful in terms of improvingsome

NSAGs’protection of subsistence rights during humanitarian crises. A human rights-specific commitment could include a provision that NSAGs who are in an ongoing conflict (or conflicts) agree that, should a situation of humanitarian crisis arise, they will fulfil certain obligations relating to subsistence rights (i.e. the minimum core). This could go some way to rectifying the main problem of using deeds to alleviate humanitarian crises—the lengthy process involved in their adoption and implementation. This approach would not be relevant for many NIAC and would still require methods capable of providing a much more immediate response. A suggested method for this will be submitted below.

Common Article 3 special agreements


Human rights agreements

Other than the voluntary commitments by NSAGs, there have also been examples of human rights agreements be-tween NSAGs and states. Perhaps the most famous of these is the human rights agreement concluded between the Frente Farabundo Martí para la Liberación Naçional (FMLN) and the government of El Salvador in 1990. The Acuerdo de San José sobre Derechos Humanos 1990 in-cluded provisions that the NSAGs would comply with the same human rights obligations as the El Salvadorian sta-te—a significant undertaking. The obligations of the FMLN in relation to particular human rights, for example the rights to freedom of association, expression and move-ment, are elaborated upon within the agreement. Al-though the more specific obligations relate more to civil and political rights as opposed to economic, social and cultural rights, the agreement was very significant for two reasons. First, in terms of disregarding concerns of states that giving NSAGs direct human rights obligations would grant the groups unwelcome legitimacy31 and second in the fact that the United Nations endorsed and agreed to monitor implementation of the agreement.32The recogni-tion of the agreement’s preamble that the FMLN had the capacity to fulfil the human rights obligations is also worthy of note, especially given widespread opposition to horizontal effect for this reason (Zegveld 2002).

Another example of a bilateral agreement including hu-man rights norms is the Comprehensive Agreement on Respect for Human Rights and International Humanitar-ian Law between the National Democratic Front of the Philippines and the Government of the Philippines 1998. The Agreement included a whole section dedicated to an impressive range of human rights.33 Although a laudable effort and very much a positive development in itself, the ensuing peace was not long-lasting. This rendered the governance potential of the agreement effectively null, with renewed efforts at reconciliation between the parties repeatedly falling through.34 In addition, the agreements would need to include specific provision for subsistence rights in order to make sure that their protection during humanitarian crises would be provided for. This may not be a problem in itself, but the low likelihood of such a hu-man rights agreement having been adopted before a situ-ation of humanitarian crisis caused by a NIAC is more problematic. Together with the rarity of states agreeing to acknowledge the validity of declarations and agreements of NSAGs, this suggests that a new approach is necessary.

Ceasefire agreements: an alternative solution? Given the above shortcomings of previous initiatives, it is submitted that a new approach must be taken to in-crease the potential of human rights observance by NSAGs during humanitarian crises. To avoid some of the challenges faced by measures such as the deeds of

commitment and more general human rights agree-ments (e.g. engagement with NSAGs and state recogni-tion of agreements), this approach should use an existing measure that is already accepted by states and NSAGs, in a new way. The measure proposed is the use of ceasefire agreements between parties to a NIAC.

The nature and content of ceasefire agreements

The term ‘ceasefire agreement’ refers to an agreement between two parties engaged in conflict with each other to end hostilities. Ceasefire agreements may take several forms, cover different scopes of content and durability and have different purposes. For example, an agreement may aim to establish peace through a complete cessation of hostilities (also referred to as ‘peace agreements’). Ceasefire agreements may also be made as a way of tem-porarily ceasing hostilities in order to enable the parties to a conflict to negotiate a full peace agreement. Alterna-tively, a ceasefire agreement may be more limited in tem-poral scope. Such agreements are often adopted for humanitarian purposes, to allow civilians temporary relief from hostilities in order to get access to essential materials (such as those seen between Hamas and Israel).35

The recommendations of using ceasefire agreements will focus on the latter type of agreement, rather than peace agreements (and long-term ceasefire agreements). The latter, although perhaps the most significant cease-fire agreements when successfully implemented, are of less relevance for present purposes. This is because al-though some such agreements have been adopted in the past (e.g. the Comprehensive Peace Accord between the Government of Nepal and the Communist Party of Nepal (Maoist) 2006), they can take many years of nego-tiation to come to fruition. Indeed, taking the example of the Colombian government, which has been negotiating various peace agreements with the FARC for decades, it is evident that in some situations, a long-term ceasefire agreement is extremely unlikely to be adopted. As the problem faced in this paper is to provide a relatively fast response to situations of humanitarian crisis, peace agree-ments would not be the most effective solution.


on which Uganda was relying in the case) (Lang 2008, 109). Andrea Lang has argued that the Court’s judge-ment was also based on a reluctance to engage with the issue of the status of NSAGs under international law (Lang 2008, 125), which is an issue that would be un-avoidable, should agreements signed by NSAGs constitute a legally binding document. However, the legal status of the kinds of ceasefire agreements between states and NSAGs focused on in the current paper remains some-what anomalous (see Kastner 2015, 13–14). Even though many of these agreements do not fall within the realm of public international law, it is still possible for a NSAG to conclude a legally binding agreement with states (for ex-ample by“including a third state party as a guarantor or using a Security Council Resolution”: Public International Law and Policy Group 2013, 7). The legal status of the agreements proposed will therefore depend upon the situ-ations of their adoption.

Ceasefire agreements generally consist of three core el-ements, which provide for “(1) a cessation of hostilities, (2) the separation of forces, and (3) the verification, supervision, and monitoring of the agreement” (Public International Law and Policy Group 2013, 1). A key component to the success of ceasefire agreements is to “clearly indicate the rights and obligations of the parties” (ibid, 1–2). This component is of particular relevance here, as it suggests that some level of detail concerning the rights and obligations is required. This supports (in light of the above discussion as to the relevantlex specia-lis) an argument that human rights norms, rather than hu-manitarian norms should be included in the agreement to improve the protection of subsistence rights.

Ceasefire agreements as a tool for compliance

In using ceasefire agreements as a tool for NSAGs’ com-pliance with human rights law to improve protection of subsistence rights during humanitarian crises, it is envis-aged that provisions detailing the relevant norms of hu-man rights law and the standards expected to be upheld by parties to the conflict will be included. The advan-tages and disadvanadvan-tages of this approach will now be examined.

Advantages of using ceasefire agreements

One advantage of using ceasefire agreements concerns the fact that their use is already widely accepted by both states and NSAGs, reflected in the prevalence of their adoption. In relation to the conflict in Myanmar alone, for example, ceasefire agreements have been adopted between the government and 14 non-state armed groups (Oo 2014, 7). In light of the reluctance of states to endorse NSAGs’ unilateral declarations or agree to Common Article 3 Spe-cial Agreements (premised on a concern that to do so would “grant a degree of legitimacy” upon the group:

ICRC 2014, 17), the acceptance by both actors is ex-tremely advantageous.

Including human rights obligations in ceasefire agree-ments that are already being negotiated would also be less resource-intensive and faster than the adoption of (for ex-ample) a new agreement specifically for the imposition of human rights obligations. In particular, short-term cease-fire agreements may be used to place economic, social and cultural rights obligations on NSAGs and states alike. The kinds of obligations envisaged would admittedly de-pend on the situation on the ground. For example, the capacity of the NSAGs to fulfil obligations and the nature of the humanitarian crisis would require careful consider-ation. However, ceasefire agreements often include hu-manitarian considerations relating to economic, social and cultural rights (for example provisions relating to the de-livery of aid).36Making the human rights aspects of these provisions explicit could be a reasonable way of placing more direct human rights obligations on NSAGs in a pos-ition to fulfil subsistence rights. This is especially true given that there are already some ceasefire agreements that include human rights-related provisions and work to-wards the protection of international humanitarian law and humanitarian aid (Reuters 2014). For example, the agreement between the Government of Nepal and the Communist Party of Nepal provides an expansive list of human rights obligations for both parties, ranging from the right to life and the prohibition of torture to the right to food and the right to health. As such, including more context-specific and detailed human rights obligations for NSAGs in ceasefire agreements would not be an excessive development.

Another advantage of using ceasefire agreements in this way is that it could open the door to more human rights agreements. Just as short-term ceasefire agree-ments often pave the way to the adoption of a peace agreement, they could form a stepping stone to more agreements like that between El Salvador and the FMLN, with a longer time span and more extensive human rights protection.


A further advantage to using ceasefire agreements is the huge symbolic value that they would have. From the perspective of NSAGs, they could be a way for them to prove that they have both the willingness and capacity to act as a state-like entity within the international arena. Whether or not this is desirable from an objective per-spective, the group may enhance its reputation both with the individuals over which they exercise control and the international community at large. They would, neverthe-less, fall short of changing the groups’status under inter-national law. Since NSAGs already conclude ceasefire agreements, no developments in the international legal framework would be necessary.

Furthermore, and of great importance, the agreements could result in less discriminatory protection of human rights. Although victims of more even and fair protec-tion of human rights would not be able to receive the sameredress for violations as they would against a state (i.e. bringing a complaint before a human rights treaty monitoring body or court), since many ceasefire agree-ments are monitored by the United Nations (which would ideally be the case here), it may be possible for them to gain some redress. The type of redress available would of course depend upon the provisions and cir-cumstances under which the agreement is drafted and the extent to which the parties would consent to be monitored. The fact that the agreements require consent from both parties would, however, increase the legitim-acy of the obligations placed on the parties.37Concerns as to the (particularly procedural) legitimacy (Ryngaert 2010, 73) of direct human rights obligations for NSAGs could be mitigated by the inclusion of the NSAG in the drafting process and ultimately by their consent in the adoption of the ceasefire agreement. This may also make NSAGs more likely to observe the obligations to which they commit themselves. Indeed, the importance of en-gaging with non-state actors before requiring certain be-haviour of them has been repeatedly stressed (see United Nations Secretary-General Ban Ki-moon 2011, cited in International Peace Institute 2012). The import-ance of engagement is reflected through the work of Geneva Call, which has“demonstrate[d] that construct-ive engagement with [armed non-state actors] can be ef-fective and can yield tangible benefits for the protection of civilians”(Bongard 2013).

Finally, from the perspectives of states, the agreements may be more palatable than the adoption of a more gen-eral agreement imposing human rights obligations on NSAGs. Indeed, in adopting an agreement, they could be seen to be fulfilling their own due diligence obliga-tions; adopting an agreement with a NSAG which would compel the group to protect human rights within the territory control could be considered a means of encour-aging NSAGs to respect human rights. In this way,

although the idea of acknowledging that NSAGs are cap-able of fulfilling some human rights obligations may not be attractive to states, doing so in a way which allows the NSAG to be held accountable may actually work in their favour.

Disadvantages of using ceasefire agreements

Whilst the inclusion of economic, social and cultural rights obligations in ceasefire agreements has many ad-vantages, they unfortunately also have some drawbacks. Most of these relate not to the use of the agreements per se, but to issues of their adoption and enforcement.

Perhaps the greatest disadvantage is the fragile nature of ceasefire agreements. Practice shows that the rate of violation of ceasefire agreements is very high. It is there-fore likely that future agreements including human rights obligations would also be breached. However, there are techniques relating to the drafting and imple-mentation of ceasefire agreements that can mitigate these risks. It has been suggested, for example, that drafting provisions within ceasefire agreements as pre-cisely as possible (in terms of the obligations and geo-graphical and temporal scope) can facilitate effective implementation (Hayson and Hottinger 2004, 2). Effect-ive implementation often relies on a monitoring mech-anism for a ceasefire agreement, agreed to between all parties (Hayson and Hottinger 2004, 6). Unfortunately, such a mechanism would be less amenable for the short-term ceasefire agreements in question. This constitutes a disadvantage of their use related to their short-term na-ture, which could prevent the agreements from having a long-term impact on economic, social and cultural rights realisation. Providing a long-term solution is not the intention behind the suggestion, however. Instead (and contrary to common ideas of seeing economic, social and cultural rights as long-term goals to be achieved in the future), one aim of the agreements is to place more focus on the potential of economic, social and cultural rights to contribute to solving very immediate problems.


disregard the international framework in place) may not consider such influences to be important (see Jo 2015). In these cases, other initiatives would have to be contem-plated. However, it is crucial to acknowledge that the pro-posed idea of including some human rights provisions in ceasefire agreements is envisaged not as a panacea. Rather, it is intended to supplement and complement existing ini-tiatives, forming one part of a multi-faceted governance solution.

The potential lack of political will of NSAGs and states alike to adopt the kind of ceasefire agreement suggested is a problem faced throughout the international human rights system. For example, the will of states to ratify hu-man rights treaties (especially relating to economic, so-cial and cultural rights) has been a challenge since their inception.39 In particular, moving from ratification as a form of lip service to the implementation of concrete human rights standards has been a constant challenge. However, the potential to have the agreements moni-tored by the United Nations, or perhaps by a different external monitoring body (perhaps even Geneva Call),40 would prove instrumental in ensuring that the obliga-tions are followed.


The above discussions aimed to outline a prevalent and persistent challenge faced during non-international armed conflicts. The humanitarian crises being suffered by so many may be mitigated to some extent through the im-position of (temporary) human rights obligations on NSAGs. As well as remedying to a degree the inequality of obligations and human rights protection, human rights norms could be used to compliment and build upon hu-manitarian obligations of all parties to non-international armed conflicts. Current initiatives that have been taken to try to alleviate the human cost of armed conflicts, in-cluding action plans, the‘list of shame’and deeds of com-mitment, have had some success in governing the behaviour of NSAGs. Nonetheless, large lacunas exist. This is particularly with regard to the ability of the initia-tives to tackle human rights protection. For this reason, a possible additional and complementary measure that could be taken is to include some subsistence rights obli-gations in short-term ceasefire agreements. This envisages giving human rights a larger role in the governance of NSAGs. Whilst not constituting a perfect solution, the agreements could prove to be a way of ameliorating situa-tions of humanitarian crisis without requiring an unrea-sonable burden on the international community.

Endnotes 1

As it is a non-governmental organisation, the defin-ition of Geneva Call is not contained in a legally binding document. However, it is very influential, given the vast

experience and work of the organisation in the field of non-international armed conflicts and in relation to NSAGs.


Universal’in the sense that each member state of the UN has ratified the Geneva Conventions. See, e.g. ICRC (2015).


This doctrine holds that since the “‘parent’state has accepted a given rule of IHL”, the state may impose the obligations upon its nationals, including those who take up arms against the state or other nationals (Kleffner 2011, 445).


That is, the argument that because individuals can be held responsible under international criminal law for war crimes, which consist of grave breaches of humani-tarian law, they must therefore be obliged to comply with humanitarian law (see Kleffner 2011, 449–451).


Article 1, para. 1 APII requires that (alongside the ex-istence of an armed conflict within the territory of a high contracting party) non-state parties to a NIAC have“ re-sponsible command, exercise such control over a part of its territory as to enable them to carry out sustained and concerted military operations and to implement this Protocol.”


See the list of inter-state vs. intra-state armed con-flicts on the International Institute for Strategic Studies (2014) Armed Conflict Database. conflicts?tags=CF582C41FE1847CF828694D51DE80C08. %20Accessed%2030%20Nov%202014. Accessed 1 Dec 2014.


Whether or not the rules are effective is another question, which falls outside of the scope of this paper.


For example, knowing how to translate the legal text into operational policies, or determining the correct scope and content of obligations.


This is evidenced by the core principle of distinction, allowing only the targeting of ‘military objectives’, but allowing any such classified individual to be killed at any time during the conflict, even when not directly partici-pating. See Article 13(1) Additional Protocol II; Solis 2010, 251–257.


Whether the killing on sight of an enemy soldier would be classified as an ‘arbitrary’ execution falls out-side the scope of this paper.


For example, Common Article 3 to the four Geneva Conventions provides for an absolute prohibition of tor-ture, of which non-observance is considered a grave breach of international humanitarian law. See, e.g. Article 130 Geneva Convention III Relative to the Treatment of Prisoners of War.


This is pursuant to the customary humanitarian rule that prohibits “attacking, destroying, removing or ren-dering useless objects indispensable to the survival of the civilian population”(see ICRC 2015a, Rule 54).





  1. (,
  4. eng/docs/v1_rul_rule55.
  19. discrimination.pdf
  21. armed-conflict.pdf
  23. geneva-call-trained-kurdish-authorities-police-forces-international-humanitarian-norms/.
  29. forces.html